Block motion vector estimation using edge matching : an approach with better frame quality as compared to full search algorithm

Abstract

Intensity based block motion estimation algorithms are widely used for exploiting temporal redundancies in video coding. However, blocks located on boundaries of moving objects are not estimated accurately, which are very sensitive to the human eyes. In this paper, we propose a novel approach which incorporates edge matching techniques to accurately predict the motion of moving objects such that the motion compensated frames are tied more closely to the physical features. Besides, accurate motion vectors of edge blocks can be used to develop an efficient block motion estimation algorithm. Experimental results show that our approach requires simple computational complexity, and it gives a significant improvement in accuracy on motion compensated frames as compared with the traditional intensity based methods, including the full search algorithm.